next »

[siskavard]

When you guys are coming up with level designs, are you sketching them out on paper first, or drawing them in photoshop (or whatever prog) or actually building them out in Gamemaker (or whatever way you build your game)

?

I'm just trying to get a feel for how to approach level design.

[TNERB]

I do all three, although building them in the game produces the best results for me. I start with a basic concept on paper and then tweak it in the actual program. Good level design takes a lot of practice, what I've been doing with my current project is making small interesting challenges in single rooms, tweaking them accordingly and then crafting a level from the better ones.

[Graham-]

@siskavard

I spend a lot of time on paper, then go to code once any element is consistent enough to make me think it's a good idea. ... Spend a lot of time bouncing between the two. There's a skill to making a level with as few elements as possible. Just use colored blocks if you have to. Add details as you become confident in their value.

The two biggest mistakes in level design:
  1. Not prototyping the key ideas soon enough, and not being minimalistic about it.
  2. Prototyping without a goal.
       -> can be as simple as, "lots of staccato jumps"

[sublinimal]

My ideal environment is a level editor that allows you to edit and playtest rapidly. Jetpack and N are good examples. I try to make a simple editor early on for my games that have clear levels.

But I also rely on pen & paper a lot. It's a better medium for thinking in concepts, puzzles, and themes rather than tiles.

Sketching in an image editor is quite rare for me, it's like the worst of both worlds.

[siskavard]

thanks for the insight, guys 

[Graham-]

Say more than that man. What is your current process? What kinds of problems do you have with it?

If you do reply make a new thread for it, then mention it here, and we'll switch over to it.

[siskavard]

my problem is I'm in the super early stages of understanding game design. I'm currently running through all of the Game Maker tutorials I can find because I'm not a coder.I figured I should be thinking about level design. All I really think about at this point is concepts for characters & enemies etc.

Once I start really trying to put something together to show I'll start a thread on it and invite everyone to bash my brain in so I can improve.

[Alec S.]

I think the main thing when level designing is to design with purpose.  Don't just say "oh, I need X number of rooms/levels/areas, I'm just going to throw increasing numbers of enemies at the player".  Starting on paper is a good way to figure out what your goal for a level is, since it's hard to lay out specifics, but easy to work in general themes and ideas. 

If the levels have linear navigation, I'll break them up into segments and give a short description of what each segment does (example), or some special aspect that area has.  It could introduce an enemy, have a combination of enemies that hasn't been used yet, or have some sort of environmental feature, for example a river or series of narrow passages, which influences gameplay.

If the level is non-liner, then you also have to think about how the player will navigate.  You still need to find ways to introduce new challenges to the player without overwhelming them by throwing a lot of things at them at once, but now they might take multiple paths through the stage.  You also have to take into account ways to guide the player through the level so that they don't feel like they don't know what to do or where to go, but they also don't feel like they're being railroaded.  For example, showing the player a treasure chest which is out of reach will motivate them to explore the area to find a way to reach it.  This would also include large-scale navigation challenges, such as key puzzles (The developer of Anodyne wrote a good article on this subject).

Once you've got your overall layout and an idea of how the player might navigate through the level, you break it down into segments and design challenges just like before (although in a more open world, some areas might be focused more on navigation than an actual challenge, you shouldn't have as much relentless action as you might have in a linear game).

Once you've gotten that done, you can then build the actual levels.  I personally think that actually building levels should wait until you've got your basic gameplay and much of your enemy/obstacle work done.  However, the level planning stage should start fairly early in the process, in parallel with designing the gameplay, enemies and such, this way you know what sort of scenarios you're building towards.  (I wrote an article talking more about this here).  Basically, you shouldn't start actually building levels until you've got the content to fill them with so you can test and adjust them right away.

And of course there's aesthetics.  Make areas look unique, give the player's nice set-pieces to have their action in, add detail layers in the tiles, give your levels interesting and pleasant to look at composition, ect...

All in all, level design is your chance to put your best foot forward.  You want to take all of your mechanics and enemies and characters and graphics and present them in the most flattering way possible.  Build scenarios around your content so that it has its chance to shine.

[TNERB]

Like Alec said always design with purpose. Pixeltao wrote a good article on level design you might wanna check out: Here.

[Graham-]

@siskavard

Ok, so you're a complete beginner. The thing you need the most is practice. ... Just try and pump out a good level, following some made up process, and play it. "Is it good?" Ask yourself this, then try again. Each time you improve your design you'll gain insight into the process.

[siskavard]

@siskavard

Ok, so you're a complete beginner. The thing you need the most is practice. ... Just try and pump out a good level, following some made up process, and play it. "Is it good?" Ask yourself this, then try again. Each time you improve your design you'll gain insight into the process.

Good advice, that's what I'm attempting right now, learning Game Maker. I want to make the most basic platformer, & not try to make this huge epic game that I'll just give up on. As an artist it's tough not to try to make it pretty, but im just using ugly blocks. If I can get the ugly blocks to do what I want I figure I'm on the right track.

[Graham-]

Yeah, simplicity and iteration. Go from concept to playable as quickly as possible, then expand that loop as you gain confidence.

[siskavard]

oh and thanks all for the links, i'm reading through them

[lolo]

Very cool thread!
This is level 9 - challenge. On the right is your shuttle, you protect it with your robot, mountable weapons and if you have enough diamonds mutagel vat or sentry. Monsters emerge from the ground in various location. The blue line you see is a scripted even, some brain bug running away. The 3 gray things in each chamber are Mutagel Towers, like a boss they shoot organic missiles that spawn more monsters. The green patches are strawberry fields the source of your economy. The pine trees boost stuff. The black part on the left is the next level that gets unlocked when you complete this one. In the next level, the brain bug runs away from a broken Glacento Tower and fortifies itself behind a force field shooting squid missiles and worms at you. The yellow blocky things with a hydrolic on top are where you can place your weapons or factories, you can also carry your weapons.

[DustyDrake]

A Doom deathmatch level I handled the visual design for, from SpaceDM9. The layout heavily emphasizes kinetic vertical movement, trick-jumping, and three-dimensional paths. I came up with the holographic cyberpunk theme for the visuals initially as a way of working around limitations of the Doom engine, but it quickly grew into something more than that.

...
*runs to dust off skulltag*

AAA! DooM is lagging! Abort, abort, abort!

[Del_Duio]

When you guys are coming up with level designs, are you sketching them out on paper first, or drawing them in photoshop (or whatever prog) or actually building them out in Gamemaker (or whatever way you build your game)

?

I'm just trying to get a feel for how to approach level design.

Weird as it is, I just kind of "wing it" a lot. I might have some ideas on how an overall area might be laid out and draw it on paper someplace but most of the time I'll see while playtesting it that this room might need a secret door here- or that two areas situated on top of one another might really benefit from some sort of connector. In a metroidvania of course these are great opportunities to have that connector be blocked off until you can open the shortcut up with a power-up or weapon you can find or buy later.

[Alec S.]

Part of a level for a rail shooter I'm working on.

[GameRoom]

One thing I've done for a platformer is make heavy emphasis on moving platforms with a robust path system. You can make a lot of different gimmicks with little addition of new elements.

[baconman]

One chunk at a time... < super-beta WIP level generator (GMK format)

Here's the EXE instead.

[7Soul]

Making a Doom2 map for MiniLD: